• Description:
  Set up Node Exporter on the target Linux server and Prometheus on a monitoring server using your OS package manager (APT). Configure Prometheus to scrape Node Exporter, verify the target is UP, and run PromQL queries to view CPU, memory and disk metrics (including useful percentage/GB calculations). Includes quick package-fix commands and basic troubleshooting steps.

•  Prelim — Fix common package issues (run only if installs fail)
   Resume interrupted package configuration:

  sudo dpkg --configure -a
        

   Fix broken dependencies:

  sudo apt install -f -y
        

   Update package list afterwards:

  sudo apt update
        

•  Step 1 — Install Node Exporter on the target server (example IP: 192.168.1.10)
   Update packages:

  sudo apt update
        

   Install Node Exporter (APT package):

  sudo apt install prometheus-node-exporter -y
        

   Start and enable the service:

  sudo systemctl start prometheus-node-exporter
  sudo systemctl enable prometheus-node-exporter
        

   Verify the service is running:

  systemctl status prometheus-node-exporter
        

   Confirm metrics are exposed:

  curl http://localhost:9100/metrics | head
        

   Expected: many metric lines including node_cpu_seconds_total, node_memory_MemAvailable_bytes, node_filesystem_avail_bytes.
  
   Node Exporter endpoint:
  http://192.168.1.10:9100/metrics
•  Step 2 — Install Prometheus on the monitoring server
   Update packages:

  sudo apt update
        

   Install Prometheus:

  sudo apt install prometheus -y
        

   Start and enable Prometheus:

  sudo systemctl start prometheus
  sudo systemctl enable prometheus
        

   Confirm service status:

  systemctl status prometheus
        

   Open Prometheus UI:
  http://<prometheus-server-ip>:9090
  (or http://localhost:9090 if local)
•  Step 3 — Configure Prometheus to scrape Node Exporter
   Edit Prometheus config:

  sudo nano /etc/prometheus/prometheus.yml
        

   Replace/add the following (adjust IPs if needed):

  global:
    scrape_interval: 15s
  
  scrape_configs:
    - job_name: 'prometheus'
      static_configs:
        - targets: ['localhost:9090']
  
    - job_name: 'node'
      static_configs:
        - targets: ['192.168.1.10:9100']
        

   Save and restart Prometheus:

  sudo systemctl restart prometheus
        

•  Step 4 — Verify Prometheus is scraping Node Exporter
   Open targets page in browser:
  http://<prometheus-server-ip>:9090/targets
  
   Confirm you see 192.168.1.10:9100 with Status = UP.
  
   If the target is DOWN, check on the target host:

  sudo systemctl status prometheus-node-exporter
  curl http://localhost:9100/metrics
        

   From Prometheus host:

  curl http://192.168.1.10:9100/metrics
        

   Check firewall / network connectivity (allow port 9100).
•  Step 5 — Run Prometheus queries (Graph tab)
   Open Prometheus UI → Graph.
  
   Run basic metric queries (paste → Execute):
  Raw CPU metric:

  node_cpu_seconds_total
        

  Available memory:

  node_memory_MemAvailable_bytes
        

  Available disk:

  node_filesystem_avail_bytes
        

   Run recommended derived queries (Execute → Graph):
  CPU usage %

  100 - (avg by (instance) (rate(node_cpu_seconds_total{mode="idle"}[5m])) * 100)
        

  Memory available %

  (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes) * 100
        

  Disk available (GB)

  node_filesystem_avail_bytes / 1024 / 1024 / 1024
        

• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
• 
•